sustainability Article A Methodological Workflow for Deriving the Association of Tourist Destinations Based on Online Travel Reviews: A Case Study of Yunnan Province, China Tao Liu 1,2, Ying Zhang 3,*, Huan Zhang 2 and Xiping Yang 4,5 1 College of Resources and Environment, Henan University of Economics and Law, Zhengzhou 450002, China; [email protected] 2 Key Laboratory of New Materials and Facilities for Rural Renewable Energy (MOA of China), Henan Agricultural University, Zhengzhou 450002, China; [email protected] 3 College of Economics and Management, Henan Agricultural University, Zhengzhou 450046, China 4 School of Geography and Tourism, Shaanxi Normal University, Xi’an 710119, China; [email protected] 5 Shaanxi Key Laboratory of Tourism Informatics, Xi’an 710119, China * Correspondence: [email protected] Abstract: Insights into the association rules of destinations can help to understand the possibility of tourists visiting a destination after having traveled from another. These insights are crucial for tourism industries to exploit strategies and travel products and offer improved services. Recently, tourism- related, user-generated content (UGC) big data have provided a great opportunity to investigate the travel behavior of tourists on an unparalleled scale. However, existing analyses of the association of destinations or attractions mainly depend on geo-tagged UGC, and only a few have utilized unstructured textual UGC (e.g., online travel reviews) to understand tourist movement patterns. In this study, we derive the association of destinations from online textual travel reviews. A workflow, Citation: Liu, T.; Zhang, Y.; Zhang, H.; Yang, X. A Methodological which includes collecting data from travel service websites, extracting destination sequences from Workflow for Deriving the travel reviews, and identifying the frequent association of destinations, is developed to achieve the Association of Tourist Destinations goal. A case study of Yunnan Province, China is implemented to verify the proposed workflow. Based on Online Travel Reviews: A The results show that the popular destinations and association of destinations could be identified in Case Study of Yunnan Province, Yunnan, demonstrating that unstructured textual online travel reviews can be used to investigate the China. Sustainability 2021, 13, 4720. frequent movement patterns of tourists. Tourism managers can use the findings to optimize travel https://doi.org/10.3390/su13094720 products and promote destination management. Academic Editor: Chia-Lin Chang Keywords: online travel review; user-generated content; association rule; movement pattern of tourist Received: 10 March 2021 Accepted: 21 April 2021 Published: 23 April 2021 1. Introduction Publisher’s Note: MDPI stays neutral Spatial movement is an essential behavior of tourism activities. Tourist movement with regard to jurisdictional claims in involves time, space, place, and scale, which are the basic elements of tourism geography. published maps and institutional affil- Tourist travel behavior can potentially imply the popularity of tourist attractions and iations. the correlation among destinations. Moreover, investigating tourist travel behavior can help uncover the intrinsic characteristics of how tourists design their itineraries, thereby helping tourism agencies and industries in planning destination facilities, assessing tourism products, and exploiting tourism resources. Therefore, tourist movement patterns have been an important research topic in tourism geography. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. Traditional approaches in investigating tourist movement patterns and destination This article is an open access article characteristics usually utilize questionnaires, but the collection of this dataset is costly distributed under the terms and and time consuming [1]. Moreover, this method is limited in sample size and space–time conditions of the Creative Commons resolution, making the analysis of tourist travel behavior from a comprehensive and broad Attribution (CC BY) license (https:// perspective difficult. Fortunately, with the rapid development of information and the creativecommons.org/licenses/by/ internet, numerous social media websites and applications (apps) allow tourists to share 4.0/). their own experiences and feelings (e.g., reviews or comments on a tourist attraction or Sustainability 2021, 13, 4720. https://doi.org/10.3390/su13094720 https://www.mdpi.com/journal/sustainability Sustainability 2021, 13, 4720 2 of 15 destination) about their travel [2–5]. These tourism-related user-generated contents (UGC) can be considered a valuable data source and open up new horizons for researchers to understand tourists’ travel experiences well and create smart urban tourism [6,7]. UGC big data can be classified into two categories: (1) geo-tagged UGC, which is produced using location-aware devices that record the location information of tourists when they post their travel experiences as comments or photos on social apps (e.g., Twitter, Flickr, and Instagram); and (2) unstructured textual content without location coordinate information on public travel service websites (e.g., TripAdvisor and Ctrip), which allows tourists to share their comments about the quality of service and release reviews of their travel experience. Geo-tagged UGC has received widespread attention from researchers in the fields of tourism, geography, and computer science because of its advantage in tracking the spatial and temporal activities of tourists [8,9]. The literature includes detecting tourism destinations or districts [10–12], characterizing tourist flows among destinations [13–15], visualizing the spatial and temporal patterns of tourists [16–19], and developing the recom- mendation model for tourist routes or attractions [20–22]. These studies show the powerful potential of geo-tagged big data in grasping insights into the spatial characteristics of tourist movement and destination correlation on an unparalleled spatial and temporal scale. Unstructured textual UGC generates a body of descriptive texts, including the com- ments or reviews of travel experiences, implying the immediate perception of tourists on destinations or travel products [3,23]. For example, these online texts can be utilized to understand destination branding or image [24–26], identify the unique or specific attribute of destinations [27,28], understand the cooperation or similarities of attractions [29,30], explore tourist movement patterns [31], and analyze tourist sentiments [32,33]. In addition, based on the comments, tourism managers will understand how tourists and customers evaluate their service (i.e., electronic word-of-mouth (eWOM)), thereby giving them ideas on how to make their management or service targeted and intelligent [34–38]. Although the second data type does not track the variation in tourist locations, it can reflect the tourists’ perception of travel activities. Nowadays, tourists tend to plan a long journey and visit more destinations with the permission of time, economy, and physical condition. Moreover, the “time–space compression” effect brought by advanced transportation expands the radiation range of tourism and makes it possible to visit more destinations during a tour on a large spatial scale; thus, multi-destination tourism has now become a popular travel mode [39]. Charac- terizing tourist movement patterns among multiple destinations will help to understand the interaction among destinations, further helping to predict the next destination [40]. Therefore, further research on multi-destination relationships is necessary. In tourism, the association rules of tourist destinations, which can be embodied from movement patterns, can quantify the possibility of tourists visiting a destination after having traveled from another. A further understanding of such rules can help predict the destinations of tourists, thus benefiting the tourism economy. In addition, the association of destinations indicates the popular destination sets and their association. Therefore, tourism practitioners and managers can use the association rules to generate targeted strategies and travel products to promote destination management and provide improved services for tourists. Currently, only a few studies have focused on tourism-related rules from UGC data. Rong et al. [41] implemented behavioral analysis on the association between web sharers and browsers and revealed the direct influence of eWOM. Based on geo-tagged UGC char- acteristics (e.g., geo-tagged photos, bluetooth tracking data), popular tourist attractions or destinations can be identified, and frequent mobility or sequential patterns can be extracted from geo-tagged travel diaries through association rule learning to understand the travel behavior and preferences of tourists [42]. In terms of data sciences in digital marketing, Saura (2021) presented a holistic overview of the framework, method, research topics, and performance metrics, and claimed that although the use of data sciences for decision- making and knowledge discovery has remarkably increased, the management of data sciences in digital marketing remained scarce [43]. Therefore, extracting knowledge from Sustainability 2021, 13, 4720 3 of 15 user-generated dataset could provide useful strategies for improving tourism management and digital marketing. At present, most of the studies only utilized the unstructured textual UGC data (e.g., online travel reviews) to understand